Researchers have assumed that the pronounced tachycardia produced by cocaine is due to sympathetic autonomic mechanisms because cocaine blocks reuptake of catecholamines. We studied the acute effects of cocaine on parasympathetic mechanisms to determine if there was a significant vagal component to this tachycardia. We used a noninvasive measure of cardiac vagal tone. This measure, which has been well-validated in animals and human subjects, is based on time series analysis of successive R - R intervals. Vagal tone index quantifies heart rate variability in the same frequency band as respiration (i.e., respiratory sinus arrhythmia). We considered vagal blockade impractical to resolve this issue because it produces very large baseline shifts in heart rate. We administered cocaine (20 mg and 40 mg) and placebo (saline) intravenously on separate days in pseudo-randomized order in double-blind fashion to 10 male residential volunteers with histories of cocaine abuse. Cocaine produced dose-dependent increases in heart rate. The effect was precisely mirrored by robust decreases in vagal tone index, as well as decreases in a lower frequency heart rate rhythm associated with blood pressure homeostasis. Injection of saline (i.e., cocaine cues) produced an initial 14 bpm increase in heart rate that had no significant vagal component. Vagal tone index and the lower frequency rhythm decreased approximately 2 to 2.5 log units in response to 40 mg cocaine, with a trough 7 to 14 min after intravenous administration. Therefore, cocaine produced a pronounced decrease in heart rate variability. The results indicate that cocaine-induced tachycardia has a strong parasympathetic component. Further research is needed to determine whether the vagal and sympathetic effects of cocaine on the autonomic nervous system are additive or synergistic. The findings have implications for a better understanding of the cardiotoxicity of cocaine.